The continuation of this grant will support the continued development of computational tools for predicting cytochrome P450 mediated regioselectivities and rates. The cytochrome P450 enzyme family is responsible for metabolism of a large number of chemicals. This enzyme family is also responsible for the bioactivation of many chemicals to reactive intermediates responsible for toxicity and carcinogenicity. Therefore, computational models for this enzyme family will play an important role in many areas of science and public health. The overall goal of this proposal is to establish computational models that can be used as a component in risk assessment. These methods will be equally applicable to the design of safer chemicals for industrial processes, as tools for drug design, and as tools for the toxicity assessment of virtual libraries of chemicals. Specifically, this grant will 1) extend existing computational models to include more P450 mediated reactions, 2) perform experiments that will allow for the prediction of the rates of P450 mediated reactions, and 3) establish 3-dimensional binding affinity models for CYP2E1, one of the most important P450 enzymes involved in the bioactivation of toxins and carcinogens. The experimental design combines the use of organic synthesis, molecular biology, protein biochemistry, analytical mass spectrometry, radiochemical methodology, and computational chemistry. Thus, this proposal also helps to train scientists in the use of multidisciplinary approaches, at the chemical/biological interface, to understanding a given scientific problem. [unreadable] [unreadable] [unreadable]